Unsupervised Multi-modal Learning

被引：2

作者：

Iqbal, Mohammed Shameer ^{[1
]}

机构：

[1] Acadia Univ, Jodrey Sch Comp Sci, Wolfville, NS B4P 2R6, Canada

来源：

ADVANCES IN ARTIFICIAL INTELLIGENCE (AI 2015) | 2015年 / 9091卷

关键词：

Unsupervised learning; Associative memory; Restricted boltzmann machines; Multi-modal data; Sensory-motor learning;

D O I：

10.1007/978-3-319-18356-5_32

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

We present an unsupervised deep belief network that can learn from multiple channels and is capable of dealing with missing information. The network learns transferable features to accommodate the addition of a new channel using a combination of unsupervised learning and a simple back-fitting.

引用

页码：343 / 346

页数：4

共 9 条

[1] SHORT-TERM SPECTRAL ANALYSIS, SYNTHESIS, AND MODIFICATION BY DISCRETE FOURIER-TRANSFORM
ALLEN, JB
[J]. IEEE TRANSACTIONS ON ACOUSTICS SPEECH AND SIGNAL PROCESSING, 1977, 25 (03): : 235 - 238
[2] Learning Deep Architectures for AI
Bengio, Yoshua
[J]. FOUNDATIONS AND TRENDS IN MACHINE LEARNING, 2009, 2 (01): : 1 - 127
[3] Geoffrey E., 2006, SCIENCE, V313, P504
[4] Hinton Geoffrey., 2010, Mo- mentum, V9
[5] Learning multiple a layers of representation
Hinton, Geoffrey E.
[J]. TRENDS IN COGNITIVE SCIENCES, 2007, 11 (10) : 428 - 434
[6] A fast learning algorithm for deep belief nets
Hinton, Geoffrey E.
Osindero, Simon
Teh, Yee-Whye
[J]. NEURAL COMPUTATION, 2006, 18 (07) : 1527 - 1554
[7] Ngiam J., 2011, PROC ICML, P689
[8] Srivastava N, 2014, J MACH LEARN RES, V15, P2949
[9] Wang T., 2014, THESIS ACADIA U

← 1 →